Multiple sclerosis (MS) is a demyelinating disease that affects 350,000 people in the U.S. and is a major cause of chronic neurological deficit, affecting adults during their most active period of their lives. Remyelination failure is a characteristic of long-standing and primary progressive lesions of MS and is associated with impulse conduction failure and axonal pathology. Despite the debilitating clinical effects of remyelination failure, the reason why some lesions are effectively remyelinated while others are not remains unclear. Glial cells that express the NG2 proteoglycan (NG2 cells) exist widely throughout the mature central nervous system. Recent genetic fate mapping studies have provided direct demonstration that they generate oligodendrocytes not only during development but also in the mature central nervous system. Using new transgenic mouse lines that we have generated, we have observed that deletion of the basic helix-loop-helix transcription factor Olig2 specifically in mature NG2 cells reduces the number of oligodendrocytes that are produced from NG2 cells in the adult brain. We have also performed a high throughput screen to identify compounds that upregulate Olig2 transcription. We will use these newly acquired tools to test the hypothesis that a critical level of Olig2 is required for successful remyelination in experimental autoimmune encephalomyelitis (EAE), which is a clinically relevant rodent model of MS. This will be tested in the following three specific aims. In Aim 1, we will determine whether loss of Olig2 will compromise the ability of NG2 cells to produce new oligodendrocytes in EAE lesions. In Aim 2, we will determine whether the newly identified compounds that increase Olig2 transcription activate the Sonic hedgehog-Gli pathway or the mitogen-activated protein kinase pathway and test compounds that affect different pathways for their ability to promote remyelination in EAE. In Aim 3, we will determine whether loss of Ezh2, which is a member of the Polycomb Repressor Complex responsible for methylating lysine 27 on histone H3, will promote remyelination by derepressing genes required for myelination.